priorsBuilder

OpenAI · released 2026-04-23

GPT-5.5

Measured on 106 benchmarks across 5 reasoning efforts.

General frontier

53.9

#12 of 32 models · best at xhigh

Reasoning efforts in your mix

xhigh53.9#43 of 129
high49.5#51 of 129
medium45.7#64 of 129
low37.9#96 of 129
none29.3#118 of 129

Your mix’s benchmarks

7 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #47xhigh of 58745.8%
  2. #49high of 58745.0%
  3. #65medium of 58742.4%
  4. #137low of 58732.7%
  5. #259none of 58713.7%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #65xhigh of 52620.5
  2. #72high of 52618.8
  3. #73medium of 52618.1
  4. #79low of 52615.1
  5. #157none of 526-4.8

Leader: Opus 5.5 · 46.4

reasoning

ARC-AGI-3

61 results
  1. #42high of 610.4%

Leader: GPT-6 Astra · 99.9%

reasoning

BullshitBench v2

196 results
  1. #82xhigh of 19647.0%
  2. #88none of 19645.0%
  3. #89low of 19645.0%
  4. #113medium of 19634.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #22xhigh of 7067.0%
  2. #29high of 7064.4%
  3. #41medium of 7054.0%
  4. #66low of 7027.0%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #17xhigh of 7072.5%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #21xhigh of 9769.0%

Leader: Opus 5.5 · 88.4%

All 106 benchmarks

show 99 more

agentic

AA-AnalystAgent

44 results
  1. #9xhigh of 4450.0%
  2. #42none of 447.5%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #95xhigh of 2241137
  2. #105high of 2241089
  3. #118medium of 2241000

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #15xhigh of 504.4%

Leader: Opus 5.5 · 15.5%

agentic

APEX-Agents-AA

33 results
  1. #6xhigh of 3337.7%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench (Zapier)

115 results
  1. #47xhigh of 11516.6%
  2. #62high of 11512.8%
  3. #79medium of 1159.0%
  4. #105low of 1152.6%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #83xhigh of 21947.3%
  2. #87high of 21944.3%
  3. #95medium of 21940.1%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #12xhigh of 5046.6%
  2. #17medium of 5044.1%

Leader: Fable 5 · 51.1%

agentic

ExploitBench v0.1

4 results
  1. #3xhigh of 447.0%

Leader: Mythos Preview · 78.0%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #40xhigh of 793.8%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #13xhigh of 4545.8%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #30xhigh of 7840.4%

Leader: Muse Spark 1.3 · 55.3%

agentic

PostTrainBench v1.1

12 results
  1. #9xhigh of 1227.2%

Leader: Fable 5 · 41.8%

agentic

SkillsBench

34 results
  1. #4xhigh of 3462.6%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #40xhigh of 6921.8%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 2.0

40 results
  1. #1xhigh of 4084.7%

Leader: this model · 84.7%

agentic

Terminal-Bench 2.1

18 results
  1. #7xhigh of 1883.2%

Leader: GPT-6 Astra · 87.4%

agentic

Time Horizon Index: KSP

13 results
  1. #10xhigh of 138.3%

Leader: Opus 5.5 · 91.3%

agentic

Vending-Bench 2

67 results
  1. #14xhigh of 67$7,524

Leader: GPT-6 Astra · $15,515

agentic

Vending-Bench Arena

12 results
  1. #3xhigh of 12$7,980

Leader: GPT-6 Astra · $12,400

agentic

τ²-Bench Telecom

393 results
  1. #38xhigh of 39393.9%
  2. #45high of 39393.0%
  3. #55medium of 39391.8%
  4. #105low of 39383.9%
  5. #160none of 39369.3%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #35xhigh of 20639.0%
  2. #43high of 20636.7%
  3. #67medium of 20629.9%
  4. #84low of 20624.9%
  5. #121none of 20614.8%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #33xhigh of 581532
  2. #36high of 581519

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #50xhigh of 1221513
  2. #58high of 1221487

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #18xhigh of 7745.2%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #1xhigh of 2970.0%
  2. #2high of 2961.9%
  3. #8medium of 2948.0%

Leader: this model · 70.0%

coding

FrontierCode 1.1 Extended

132 results
  1. #53xhigh of 13256.7%
  2. #73high of 13254.0%
  3. #83medium of 13251.4%
  4. #100low of 13245.6%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

132 results
  1. #55xhigh of 13243.0%
  2. #69high of 13240.6%
  3. #86medium of 13236.6%
  4. #100low of 13230.6%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v1

17 results
  1. #7xhigh of 1764.5%

Leader: Fable 5 · 88.2%

coding

LiveCodeBench (Vals)

127 results
  1. #35xhigh of 12785.3%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #29xhigh of 629.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #23xhigh of 620.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #20xhigh of 6270.8%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #40high of 22656.1%
  2. #43xhigh of 22655.8%
  3. #59medium of 22654.5%

Leader: Opus 5.5 · 66.9%

coding

SWE Atlas-QnA

22 results
  1. #8xhigh of 2245.4%

Leader: Opus 5 · 63.2%

coding

SWE-bench Verified (Vals)

85 results
  1. #18xhigh of 8582.6%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #1high of 6373.2%

Leader: this model · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #16high of 7476.4%
  2. #43xhigh of 7457.3%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #7xhigh of 38760.6%
  2. #9high of 38759.8%
  3. #14medium of 38757.6%
  4. #19low of 38752.3%
  5. #24none of 38749.2%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #33xhigh of 23884.3%
  2. #43medium of 23880.5%
  3. #48high of 23879.4%
  4. #80low of 23865.5%
  5. #97none of 23861.0%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #69xhigh of 21614.6%
  2. #88high of 2169.1%
  3. #98medium of 2165.1%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #39xhigh of 10569.8%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #14xhigh of 12784.9%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #73xhigh of 61038.4
  2. #80high of 61037.0
  3. #96medium of 61033.8
  4. #113low of 61030.7
  5. #186none of 61023.2

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #23high of 2201482
  2. #28xhigh of 2201477

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #21high of 2201487
  2. #30xhigh of 2201484

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #9xhigh of 3368.1%

Leader: Fable 5 · 74.2%

composite

Vals RSI Index v1.1

24 results
  1. #20xhigh of 2416.5%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #22xhigh of 245.6%

Leader: GPT-6 Sol · 22.8%

composite

Vals RSI Index v1.1: Post-training: Finance Agent

24 results
  1. #23xhigh of 240.0%

Leader: Opus 5.5 · 35.1%

composite

Vals RSI Index v1.1: Pre-training: Compression

24 results
  1. #16xhigh of 2429.2%

Leader: Opus 5.5 · 46.5%

composite

Vals RSI Index v1.1: Pre-training: LM Training

24 results
  1. #19xhigh of 2431.1%

Leader: Opus 5.5 · 53.0%

cost

AA Intelligence Index Cost per Task

177 results
  1. #109medium of 177$0.90
  2. #137high of 177$1.54
  3. #157xhigh of 177$2.63

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #137medium of 177$2,095
  2. #153high of 177$3,411
  3. #166xhigh of 177$5,294

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #11xhigh of 721883

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (image)

16 results
  1. #3xhigh of 161527

Leader: GPT-6.1 Sol · 2204

creative

VoxelBench (text)

55 results
  1. #10xhigh of 551975

Leader: GPT-6 Astra · 2733

economics

CorpFin (Vals)

117 results
  1. #9xhigh of 11768.4%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #25xhigh of 7464.5%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #37xhigh of 7851.8%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #48high of 21022.6%
  2. #54xhigh of 21021.2%
  3. #60medium of 21020.6%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #86xhigh of 2881350
  2. #91high of 2881327
  3. #112medium of 2881231
  4. #149low of 2881044
  5. #166none of 288969

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #10xhigh of 8468.8%

Leader: Opus 5 · 72.1%

economics

Public Benefits Bench v1 (Vals)

13 results
  1. #8xhigh of 1357.2%

Leader: Fable 5 · 71.7%

economics

TaxEval (Vals)

125 results
  1. #25xhigh of 12575.0%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #26xhigh of 39975.9%
  2. #51high of 39971.6%
  3. #62medium of 39971.0%
  4. #109low of 39964.4%
  5. #195none of 39946.1%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #28xhigh of 52658.0%
  2. #32high of 52657.0%
  3. #33medium of 52656.8%
  4. #37low of 52654.9%
  5. #85none of 52645.6%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #369low of 52612.0%
  2. #385xhigh of 52611.0%
  3. #386high of 52610.9%
  4. #393medium of 52610.4%
  5. #461none of 5267.3%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

Arena Search

28 results
  1. #3xhigh of 281242

Leader: GPT-5.6 Sol · 1257

knowledge

CaseLaw (Vals)

53 results
  1. #7xhigh of 5366.2%

Leader: Grok 4.3 · 79.3%

knowledge

GPQA Diamond

539 results
  1. #20xhigh of 53993.5%
  2. #23high of 53993.2%
  3. #34medium of 53992.6%
  4. #54low of 53991.0%
  5. #224none of 53976.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #11xhigh of 12293.2%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #26xhigh of 9449.1%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #19xhigh of 9686.9%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #21xhigh of 12288.1%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #9xhigh of 8188.3%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #39medium of 26581.2%
  2. #40high of 26581.1%
  3. #52xhigh of 26579.9%
  4. #61low of 26579.0%
  5. #133none of 26571.4%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #32xhigh of 5060.9%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #9xhigh of 52384.3%
  2. #10high of 52384.3%
  3. #32medium of 52383.0%
  4. #60low of 52381.0%
  5. #252none of 52364.0%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #8xhigh of 431486
  2. #9high of 431484

Leader: Opus 5 · 1516

long-context

MLCR-AA

94 results
  1. #37xhigh of 9417.8%
  2. #52none of 9415.0%
  3. #62low of 9411.1%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Erdős

7 results
  1. #5xhigh of 70.0%

Leader: GPT-6 Astra · 2.9%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #12xhigh of 11585.3%

Leader: GPT-6.1 Sol · 93.7%

other

Arena Vision

116 results
  1. #19xhigh of 1161286
  2. #24high of 1161281

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #16xhigh of 850.1%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #33xhigh of 21595.0%
  2. #36high of 21594.5%
  3. #52medium of 21592.2%
  4. #114low of 21576.2%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #28xhigh of 21685.0%
  2. #35high of 21683.3%
  3. #51medium of 21670.4%
  4. #111low of 21633.3%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #9xhigh of 2836.2%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #34xhigh of 52927.1%
  2. #43high of 52925.4%
  3. #66medium of 52918.6%
  4. #131low of 5298.0%
  5. #206none of 5291.4%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Furniture Assembly

31 results
  1. #12xhigh of 3144.2%

Leader: Opus 5.5 · 83.3%

reasoning

LegalBench

135 results
  1. #12xhigh of 13586.5%

Leader: Fable 5 · 88.6%

reasoning

Roboflow Visual Reasoning

102 results
  1. #34high of 10272.2%
  2. #39low of 10270.6%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #13xhigh of 8051.5%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

172 results
  1. #40xhigh of 1720.4%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #70xhigh of 18299.8%

Leader: GPT-5 Pro · 100.0%

safety

SRE Bench

32 results
  1. #18xhigh of 323.8%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

351 results
  1. #49none of 3510.9s
  2. #111low of 3511.5s
  3. #246medium of 3519.3s
  4. #304high of 35137s
  5. #323xhigh of 35171s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #157xhigh of 351104.9
  2. #169high of 35198.5
  3. #185medium of 35191.9
  4. #192none of 35188.1
  5. #194low of 35186.2

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #150xhigh of 1783.3s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #97xhigh of 17657.5

Leader: Muse Spark 1.1 · 204.5